CN113722601A - Power measurement information recommendation method and device, computer equipment and storage medium - Google Patents

Power measurement information recommendation method and device, computer equipment and storage medium Download PDF

Info

Publication number
CN113722601A
CN113722601A CN202111043138.6A CN202111043138A CN113722601A CN 113722601 A CN113722601 A CN 113722601A CN 202111043138 A CN202111043138 A CN 202111043138A CN 113722601 A CN113722601 A CN 113722601A
Authority
CN
China
Prior art keywords
information
user
power measurement
measurement information
target
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202111043138.6A
Other languages
Chinese (zh)
Other versions
CN113722601B (en
Inventor
杨劲锋
龚起航
郑楷洪
李胜
周尚礼
曾璐琨
刘玉仙
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Southern Power Grid Digital Grid Research Institute Co Ltd
Original Assignee
China Southern Power Grid Co Ltd
Southern Power Grid Digital Grid Research Institute Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by China Southern Power Grid Co Ltd, Southern Power Grid Digital Grid Research Institute Co Ltd filed Critical China Southern Power Grid Co Ltd
Priority to CN202111043138.6A priority Critical patent/CN113722601B/en
Publication of CN113722601A publication Critical patent/CN113722601A/en
Application granted granted Critical
Publication of CN113722601B publication Critical patent/CN113722601B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/953—Querying, e.g. by the use of web search engines
    • G06F16/9535—Search customisation based on user profiles and personalisation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/335—Filtering based on additional data, e.g. user or group profiles
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/048—Activation functions
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0281—Customer communication at a business location, e.g. providing product or service information, consulting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Electricity, gas or water supply

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • Development Economics (AREA)
  • General Health & Medical Sciences (AREA)
  • Human Resources & Organizations (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • Tourism & Hospitality (AREA)
  • Biophysics (AREA)
  • Software Systems (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Mathematical Physics (AREA)
  • Computing Systems (AREA)
  • Game Theory and Decision Science (AREA)
  • Molecular Biology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Biomedical Technology (AREA)
  • Evolutionary Computation (AREA)
  • Quality & Reliability (AREA)
  • Primary Health Care (AREA)
  • Water Supply & Treatment (AREA)
  • Public Health (AREA)
  • Operations Research (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The application relates to a power measurement information recommendation method and device, computer equipment and a storage medium. When a user to be recommended triggers a power measurement information consultation instruction, a target user database is inquired by using a user identifier of the user to be recommended and user consultation information, the obtained target user information is input into a target power measurement information recommendation model, target power measurement information corresponding to the user to be recommended and output by the target power measurement information recommendation model is obtained, and the output target power measurement information is displayed. Compared with the current power measurement information recommendation which can not be carried out according to the user requirements according to the characteristics of the user, the scheme realizes reasonable and accurate power measurement information recommendation for the user through a recommendation model based on the user identity information, the user consultation information and the target power measurement information.

Description

Power measurement information recommendation method and device, computer equipment and storage medium
Technical Field
The present application relates to the field of machine learning technologies, and in particular, to a power measurement information recommendation method and apparatus, a computer device, and a storage medium.
Background
In the field of electric power, a user can generate a lot of electric power measurement information in the electricity utilization process and the operation process of electric power equipment, and the electric power measurement information has a high reference value for the user, and if the electric power measurement information can be effectively utilized, help can be brought to the life and social production development of the user. However, in the power field, there is still a lack of a way to make a reasonable recommendation of power measurement data for users.
Therefore, how to reasonably and accurately recommend the power measurement information to the user becomes a problem which needs to be solved urgently.
Disclosure of Invention
In view of the above, it is necessary to provide a power measurement information recommendation method, device, computer device and storage medium, which can implement power measurement information recommendation to users reasonably and accurately.
A power measurement information recommendation method, the method comprising:
responding to a power measurement information consultation instruction triggered by a user to be recommended, and acquiring a user identifier and user consultation information of the user to be recommended;
inquiring a target user database according to the user identification and the user consultation information to acquire corresponding target user information; the target user information comprises user identity information and historical consultation information corresponding to the user to be recommended; the target user database stores corresponding relations between a plurality of user identifications and historical consultation information;
inputting the target user information into a target power measurement information recommendation model, and acquiring and displaying target power measurement information corresponding to the user to be recommended and output by the target power measurement information recommendation model; the target power measurement information recommendation model is obtained by training based on a plurality of sample user identity information, a plurality of historical consultation information and a plurality of sample power measurement information.
In one embodiment, the method further comprises:
acquiring a historical power measurement information consultation record corresponding to a user;
acquiring user identity information of the user, historical consultation questions of the user and corresponding historical consultation answers from the historical electric power measurement information consultation record;
identifying the historical consultation question and the power measurement information in the historical consultation response as historical consultation information;
and obtaining the user database according to the user identity information and the historical consultation information.
In one embodiment, the method further comprises:
acquiring a plurality of sample user identity information, a plurality of historical consultation information and a plurality of sample power measurement information; a set of the sample user identity information and historical consulting information corresponds to at least one of the sample power measurement information;
acquiring a first power measurement information recommendation model to be trained, inputting the sample user identity information and the corresponding historical consultation information into the first power measurement information recommendation model, acquiring predicted power measurement information output by the first power measurement information recommendation model, and judging whether the similarity between the predicted power measurement information and the sample user identity information as well as the similarity between the predicted power measurement information and the sample power measurement information corresponding to the historical consultation information is greater than or equal to a preset similarity threshold value;
if not, adjusting the first power measurement information recommendation model according to the similarity, and returning to the step of inputting the sample user identity information and the corresponding historical consultation information into the first power measurement information recommendation model;
if so, ending the circulation, and taking the current first power measurement information recommendation model as the target power measurement information recommendation model.
In one embodiment, the obtaining of the first power measurement information recommendation model to be trained includes:
constructing an embedding layer; the embedding layer is used for converting the sample user identity information into a first low-dimensional vector and converting the historical consultation information into a second low-dimensional vector;
constructing an attention layer; the attention layer is used for identifying power measurement information in the first low-dimensional vector and the second low-dimensional vector through a multi-head attention mechanism;
constructing a multilayer perceptron; the multilayer perceptron is used for identifying whether the power measurement information output by the attention layer is power measurement information corresponding to the identity of a sample user, taking the power measurement information corresponding to the identity of the sample user as predicted power measurement information and outputting the predicted power measurement information through a preset activation function;
and obtaining the first power measurement information recommendation model according to the embedded layer, the attention layer and the multilayer perceptron.
In one embodiment, the inputting the sample user identity information and the corresponding historical consulting information into the first power measurement information recommendation model includes:
converting the sample user identity information into a first word vector and converting the historical consulting information into a second word vector;
inputting the first word vector and the second word vector into the first power measurement information recommendation model.
In one embodiment, the querying a target user database according to the user identifier and the user consultation information to obtain corresponding target user information includes:
if the user database has user identity information corresponding to the user identification, obtaining the target user information according to the user identity information corresponding to the user identification and historical consultation information corresponding to the user identity information;
if the user database does not have the user identity information corresponding to the user identification, obtaining the most similar historical consultation information in the user database to the user consultation information, and obtaining the target user information according to the most similar historical consultation information and the user identity information corresponding to the most similar historical consultation information.
In one embodiment, the inputting the target user information into a target power measurement information recommendation model, and acquiring and displaying target power measurement information corresponding to the user to be recommended and output by the target power measurement information recommendation model includes:
inputting the user identity information and the historical consultation information into a target power measurement information recommendation model;
acquiring a plurality of target power measurement information corresponding to the user to be recommended and output by the target power measurement information model and the prediction accuracy corresponding to each target power measurement information;
and displaying a preset number of target power measurement information in the plurality of target power measurement information according to the prediction accuracy.
An electricity measurement information recommendation device, the device comprising:
the acquisition module is used for responding to a power measurement information consultation instruction triggered by a user to be recommended, and acquiring a user identifier and user consultation information of the user to be recommended;
the query module is used for querying a target user database according to the user identification and the user consultation information to acquire corresponding target user information; the target user information comprises user identity information and historical consultation information corresponding to the user to be recommended; the target user database stores corresponding relations between a plurality of user identifications and historical consultation information;
the recommendation module is used for inputting the target user information into a target power measurement information recommendation model, and acquiring and displaying target power measurement information corresponding to the user to be recommended and output by the target power measurement information recommendation model; the target power measurement information recommendation model is obtained by training based on a plurality of sample user identity information, a plurality of historical consultation information and a plurality of sample power measurement information.
A computer device comprising a memory storing a computer program and a processor implementing the steps of the method described above when executing the computer program.
A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the above-mentioned method.
According to the power measurement information recommendation method, device, computer equipment and storage medium, when the power measurement information consultation instruction is triggered by the user to be recommended, the user identification and the user consultation information of the user to be recommended are used for inquiring the target user database, the obtained target user information is input into the target power measurement information recommendation model, the target power measurement information corresponding to the user to be recommended and output by the target power measurement information recommendation model is obtained, and the output target power measurement information is displayed. Compared with the current power measurement information recommendation which can not be carried out according to the user requirements according to the characteristics of the user, the scheme realizes reasonable and accurate power measurement information recommendation for the user through a recommendation model based on the user identity information, the user consultation information and the target power measurement information.
Drawings
FIG. 1 is a diagram of an exemplary power measurement information recommendation method;
FIG. 2 is a flow chart illustrating a method for recommending power measurement information according to an embodiment;
FIG. 3 is a flowchart illustrating the training steps of the target power measurement information recommendation model in one embodiment;
FIG. 4 is a schematic flow chart diagram illustrating the attention layer construction step in one embodiment;
FIG. 5 is a flowchart illustrating a method for recommending power measurement information according to another embodiment;
FIG. 6 is a flow chart illustrating a power measurement recommendation method according to yet another embodiment;
FIG. 7 is a block diagram of an embodiment of an electricity measurement information recommendation device;
FIG. 8 is a diagram illustrating an internal structure of a computer device according to an embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
The power measurement information recommendation method provided by the application can be applied to the application environment shown in fig. 1. The terminal 102 can acquire the user identifier and the user consultation information of the user to be recommended when detecting the power measurement information consultation instruction triggered by the recommending user, and query the target user database according to the user identifier and the user consultation information to acquire the corresponding target user information, so that the terminal 102 can display the target power measurement information corresponding to the user to be recommended according to the target user information and the target power measurement information recommendation model, and the power measurement information can be reasonably and accurately recommended to the user to be recommended. Among other things, in some embodiments, a server 104 may also be included. Wherein the terminal 102 communicates with the server 104 via a network. The terminal 102 may set the target user database in the server 104, so that the terminal 102 may send a query instruction to the server 104, so that the server 104 may query the target user information and return the target user information to the terminal 102. The terminal 102 may be, but is not limited to, various personal computers, notebook computers, smart phones, and tablet computers, and the server 104 may be implemented by an independent server or a server cluster formed by a plurality of servers.
In an embodiment, as shown in fig. 2, a power measurement information recommendation method is provided, which is described by taking the method as an example applied to the terminal in fig. 1, and includes the following steps:
step S202, responding to a power measurement information consultation instruction triggered by a user to be recommended, and acquiring a user identification and user consultation information of the user to be recommended.
The user to be recommended may be a user who needs to recommend the power measurement information. The user to be recommended may trigger the power measurement information consultation instruction in the terminal 102, for example, after the user to be recommended inputs the user identifier and the user consultation information to be consulted in the terminal 102, the user to be recommended clicks a corresponding button in the terminal 102 to trigger the consultation. When the terminal 102 detects the power measurement information consultation instruction, the user identification and the user consultation information of the user to be recommended can be obtained, so that the terminal 102 can obtain the power measurement information meeting the requirements of the user to be recommended based on the user identification and the user consultation information and recommend the power measurement information to the user to be recommended.
Step S204, inquiring a target user database according to the user identification and the user consultation information to acquire corresponding target user information; the target user information comprises user identity information and historical consultation information corresponding to the user to be recommended; the target user database stores the corresponding relation between a plurality of user identifications and historical consultation information.
The user identifier may be an identifier obtained by the terminal 102 through the camera device to identify the user, or an identifier input by the user; the user consultation information may be consultation information of the user to be recommended, which is input into the terminal 102 in the power measurement field, for example, the terminal 102 may obtain the user consultation information through a voice receiving device, and may also obtain the user consultation information through a text receiving device. A target user database may also be provided in the terminal 102. The target user database may store a corresponding relationship between a plurality of user identifiers and historical consulting information. After the terminal 102 acquires the user identifier and the user consultation information of the user to be recommended, the terminal 102 can query the target user database through the user identifier and the user consultation information, so that the terminal 102 can query whether the user identity information and the historical consultation information corresponding to the user identifier or the user consultation information exist in the target user database, and when the terminal 102 queries the user identity information and the historical consultation information corresponding to the user identifier or the user consultation information, the queried user identity information and the queried historical consultation information can be acquired and serve as the target user information.
Step S206, inputting the information of the target user into a target power measurement information recommendation model, and acquiring and displaying target power measurement information corresponding to the user to be recommended and output by the target power measurement information recommendation model; the target power measurement information recommendation model is obtained based on a plurality of sample user identity information, a plurality of historical consultation information and a plurality of sample power measurement information training.
The target user information may include user identity information, historical consulting information and the like of the user to be recommended. The terminal 102 may drive the acquired target user information into the target power measurement information recommendation model, the target power measurement information recommendation model may acquire and output target power measurement information interested by the user to be recommended according to the input target user information, and the terminal 102 may acquire the target power measurement information output by the target power measurement information recommendation model as the power measurement information interested by the user to be recommended. The power measurement information may be knowledge in the power measurement field, and the target power measurement information recommendation model may be obtained through training, for example, training based on the plurality of sample user identity information, the plurality of historical consultation information, and the plurality of sample power measurement information. The sample user identity information and the historical consulting information can be obtained from a user database, the sample power measurement information can be obtained from a power measurement knowledge base, and the power measurement information recommendation model can be a deep learning model.
The terminal 102 may also display the target power measurement information obtained through the target power measurement information recommendation model, for example, through a display device in the terminal 102, so that the user to be recommended may visually obtain the target power measurement information. The obtained target power measurement information may include a plurality of pieces of information, so that the target power measurement information displayed by the terminal 102 may also be a plurality of pieces of information. For example, in one embodiment, inputting the target user information into the target power measurement information recommendation model, and acquiring and displaying the target power measurement information corresponding to the user to be recommended and output by the target power measurement information recommendation model includes: inputting user identity information and historical consultation information into a target power measurement information recommendation model; acquiring a plurality of target power measurement information corresponding to a user to be recommended and output by a target power measurement information model and the prediction accuracy corresponding to each target power measurement information; and displaying a preset number of target power measurement information in the plurality of target power measurement information according to the prediction accuracy. In this embodiment, the terminal 102 may input the user identity information and the historical consulting information in the target user information into the target power measurement information recommendation model, and the target power measurement information recommendation model may obtain and output a plurality of target power measurement information and the prediction accuracy corresponding to each target power measurement information according to the user identity information and the historical consulting information, where the prediction accuracy may represent the interest degree of the user to be recommended in each target power measurement information. Therefore, the terminal 102 displays the target power measurement information of the preset number in the plurality of target power measurement information according to the prediction accuracy of the item mark power measurement information. For example, the power measurement information may be power measurement knowledge, and the terminal 102 may obtain the prediction accuracy of each of the plurality of power measurement knowledge output by the model, and select a preset number of power measurement knowledge from the plurality of power measurement knowledge according to the number required to be displayed according to the number of the prediction accuracy from large to small, for example, the terminal 102 selects five power measurement knowledge from large to small according to the prediction accuracy to display. Thereby completing the recommendation of the power measurement knowledge.
According to the power measurement information recommendation method, when a power measurement information consultation instruction is triggered by a user to be recommended, a target user database is inquired by using a user identifier of the user to be recommended and user consultation information, the obtained target user information is input into a target power measurement information recommendation model, target power measurement information corresponding to the user to be recommended and output by the target power measurement information recommendation model is obtained, and the output target power measurement information is displayed. Compared with the current power measurement information recommendation which can not be carried out according to the user requirements according to the characteristics of the user, the scheme realizes reasonable and accurate power measurement information recommendation for the user through a recommendation model based on the user identity information, the user consultation information and the target power measurement information.
In one embodiment, further comprising: acquiring a historical power measurement information consultation record corresponding to a user; acquiring user identity information of a user, historical consultation questions of the user and corresponding historical consultation answers from historical electric power measurement information consultation records; identifying power measurement information in the historical consultation questions and the historical consultation answers as historical consultation information; and obtaining a user database according to the user identity information and the historical consultation information.
In this embodiment, the terminal 102 may construct the user database based on the personal information of the user and the consultation information of the user on the power measurement. The terminal 102 may obtain a historical power measurement information consultation record of the user, and obtain user identity information of the user, historical consultation questions of the user, and historical consultation responses corresponding to the historical consultation questions from the historical power measurement information consultation record, and the terminal 102 may identify the power measurement information from the historical consultation questions and the corresponding historical consultation responses, for example, extract information by identifying a keyword, and use the identified power measurement information as the historical consultation information; so that the terminal 102 can form a user database according to the plurality of user identity information and the corresponding historical consultation information.
The user identity information may include information such as a mobile phone number and a name of the user, and the terminal 102 may bind and store the information of the mobile phone number and the name of the user and the historical consultation information corresponding to the user. For example, in the process of constructing the user database, the terminal 102 may count the user information of the questions related to the consulted power consumption, and collect the user information, the questions consulted by the user and the corresponding responses of the customer service in the service record of the manual customer service chat system of the power enterprise; user information, questions consulted by the user and corresponding responses of the customer service can be collected in the service record of the telephone customer service of the power enterprise; the terminal 102 may screen out the data related to the power measurement in the collected data, such as "power consumption", "voltage", and the like; taking the mobile phone number or name of each user as a main key, and recording the consulted questions related to the power measurement data into the information record of the user; and establishing a data record containing the questions consulted by each consulted user to form a user information database.
Through the embodiment, the terminal 102 can extract the power measurement information by using data generated in the process of consulting the power measurement information by a plurality of users, and the power measurement information and the personal information of the users are stored together to form the user database, so that the terminal 102 can recommend the power measurement information to the user to be recommended based on the user database, and reasonable and accurate power measurement information recommendation is realized.
In one embodiment, further comprising: acquiring a plurality of sample user identity information, a plurality of historical consultation information and a plurality of sample power measurement information; a group of sample user identity information and historical consulting information correspond to at least one sample power measurement information; acquiring a first power measurement information recommendation model to be trained, inputting sample user identity information and corresponding historical consultation information into the first power measurement information recommendation model, acquiring predicted power measurement information output by the first power measurement information recommendation model, and judging whether the similarity between the predicted power measurement information and the sample user identity information and the similarity between the predicted power measurement information and the sample power measurement information corresponding to the historical consultation information are larger than or equal to a preset similarity threshold value or not; if not, adjusting the first power measurement information recommendation model according to the similarity, and returning to the step of inputting the sample user identity information and the corresponding historical consultation information into the first power measurement information recommendation model; if so, ending the circulation, and taking the current first power measurement information recommendation model as a target power measurement information recommendation model.
In this embodiment, the terminal 102 may train the power measurement information recommendation model according to the identity information of the user, the historical consultation information corresponding to the identity information of the user, and the sample power measurement information. The terminal 102 may obtain a plurality of sample user identity information, a plurality of historical consulting information and a plurality of sample power measurement information, and a set of sample user identity information and historical consulting information corresponds to at least one sample power measurement information. That is, a set of sample subscriber identity information and historical advisory information corresponds to one or more sample power measurement information. The terminal 102 may obtain a first power measurement information recommendation model to be trained, input the sample user identity information and corresponding historical consultation information into the first power measurement information recommendation model, and obtain predicted power measurement information output by the first power measurement information recommendation model, so that the terminal 102 may determine whether the similarity between the predicted power measurement information and the sample user identity information and the similarity between the predicted power measurement information and the sample power measurement information corresponding to the historical consultation information are greater than or equal to a preset similarity threshold; if the terminal 102 determines that the similarity between the predicted power measurement information and the corresponding sample power measurement information is smaller than the preset similarity threshold, the terminal 102 may adjust the first power measurement information recommendation model according to the similarity, select new sample user identity information, historical consultation information and corresponding sample power measurement information from the plurality of sample user identity information, the plurality of historical consultation information and the plurality of sample power measurement information, return to the step of inputting the sample user identity information and the corresponding historical consultation information into the first power measurement information recommendation model, and perform the next training. When the terminal 102 determines that the similarity is greater than or equal to the preset similarity threshold, the terminal 102 may end the loop, and use the current first power measurement information recommendation model as the target power measurement information recommendation model, thereby completing training of the target power measurement information recommendation model.
The sample power measurement information may be obtained from a power measurement knowledge base, and the power measurement knowledge base may store a plurality of power measurement information. The terminal 102 may first build a power measurement knowledge base. In the process of constructing the power measurement knowledge base, the terminal 102 can acquire relevant data from a database inside a power grid enterprise; acquiring power related data from public media, such as collecting related data information from government websites, power enterprise websites and power industry periodicals; processing such as duplicate removal, alignment and the like is carried out on the collected data; determining the names of the measurement knowledge to be recorded in the knowledge base, such as data of 'power consumption amount', 'voltage', and the like; and storing the sorted data into a database to form a formal power measurement knowledge base. The terminal 102 may thus train the first power measurement information recommendation model using the power measurement knowledge base and the data in the user database. For example, as shown in fig. 3, fig. 3 is a flowchart illustrating a training step of the target power measurement information recommendation model in one embodiment. The terminal 102 may obtain a first power measurement information recommendation model to be trained by constructing a multi-layer network, including an Embedding layer, a transform layer, and an MLP (multi-layer perceptron) layer. The terminal 102 may build a first power measurement information recommendation model using the above layers. For example, in one embodiment, obtaining a first power measurement information recommendation model to be trained includes: constructing an embedding layer; the embedded layer is used for converting the sample user identity information into a first low-dimensional vector and converting the historical consultation information into a second low-dimensional vector; constructing an attention layer; the attention layer is used for identifying power measurement information in the first low-dimensional vector and the second low-dimensional vector through a multi-head attention mechanism; constructing a multilayer perceptron; the multi-layer sensing machine is used for identifying whether the power measurement information output by the attention layer is the power measurement information corresponding to the identity of the sample user, taking the power measurement information corresponding to the identity of the sample user as the predicted power measurement information and outputting the predicted power measurement information through a preset activation function; and obtaining a first power measurement information recommendation model according to the embedded layer, the attention layer and the multilayer perceptron. In this embodiment, the terminal 102 may first construct an embedding layer in the deep learning model, where the embedding layer may be configured to convert sample user identity information into a first low-dimensional vector and convert corresponding historical consultation information into a second low-dimensional vector, that is, embed the features of the power measurement knowledge, the user information features, and each combination feature into a low-dimensional vector; the terminal 102 may also construct an attention layer, that is, construct a Transformer layer, as shown in fig. 4, where fig. 4 is a flowchart illustrating an attention layer constructing step in an embodiment. The Transformer layer may be configured to learn deeper characterization relationships between the power measurement knowledge features or the user information features and other data, and the terminal 102 may identify the power measurement information in the first low-dimensional vector and the second low-dimensional vector through a multi-head attention mechanism in the attention layer. The transform layer may include a Self-Attention layer and an FNN (feedforward-Neural network) module; the terminal 102 may use each module in the transform layer to learn the characterization relationship. The terminal 102 may further construct a multi-Layer aware (MLP), which may be configured to identify whether the power measurement information output by the attention Layer is power measurement information corresponding to the identity of the sample user, and output the power measurement information corresponding to the identity of the sample user as predicted power measurement information through a preset activation function; namely, the MLP layer can use three full-connection layers to further learn the cross feature information among various dense features, and a sigmoid activation function is used as an output unit. The terminal 102 may obtain the first power measurement information recommendation model according to the embedded layer, the attention layer, and the multi-layer perceptron. So that the terminal 102 may train the first power measurement information recommendation model constructed as described above.
For example, the terminal 102 may input all the characteristics of the power measurement data and the user information data into a low-dimensional vector with a fixed size by using the embed layer; embedding the user information and the power measurement data position characteristics as two main characteristics; the terminal 102 may embed the user information in: user name and user mobile phone numberThe data of the two fields are used as main data for embedding the user information; the terminal 102 may also embed the power measurement data location: capturing order information in a sentence using position embedding; therefore, the terminal 102 can combine the user information embedding and the power measurement data position embedding to generate an embedded matrix Wv∈R(|V|×dv). Wherein d isvFor embedded dimension size, | V | is the amount of data. For the Self-orientation layer, the orientation in the transform is defined as:
Figure BDA0003250155300000111
where Q represents a query, K represents a key, and V represents a value. The terminal 102 may use a multi-head attention mechanism to improve its ability to extract data context information; the terminal 102 may use the embedded matrix WVAs input and converted into three matrices, which are input into the multi-head attention layer. For the feedforward neural network module FNN, the terminal 102 may add a feedforward data network in the transform layer to enhance the nonlinear capability of the model, which is defined as: f ═ ffn(s); to avoid overfitting and learn more meaningful features from depth, the terminal 102 may use Dropout and LeakyReLU activation functions in the Self-Attention layer (Self-Attention) and feed-forward neural network modules (FFNs) to discard some meaningless feature neurons and to optimize some neurons that are accidentally lost. The overall output of Self-Attention and FFN is as follows:
S′＝LayerNorm(S+Dropout(MH(S)))；
F＝LayerNorm(S′+Dropout(Leaky Re LU(S′W(1)+b(1))W(2)+b(2)))；
wherein, W(1)、b(1)、W(2)And b(2)Are all learnable parameters. For a multi-layer perceptron, the terminal 102 may learn cross information between features by connecting outputs of the transform layer using three full-connection layers, each full-connection layer using a LeakyReLU activation function; the terminal 102 may set the problem of the recommendation prediction to twoClassification problems, such as classification into user interest and non-interest in a certain power measurement data information; and uses the Sigmoid function as an output unit. The terminal 102 may thus train the first power measurement information recommendation model using the above layers.
Through the embodiment, the terminal 102 may use a deep learning model with a multilayer structure to train the model based on the user identity information, the historical consultation information and the sample power measurement information, so as to obtain a target power measurement information recommendation model that may be used for power measurement information recommendation. The terminal 102 combines the Transformer model which has excellent effect in the natural language processing field into the whole model architecture, and the accuracy of knowledge recommendation is effectively improved. The electric power measurement information recommendation method and the electric power measurement information recommendation device can accurately and reasonably recommend electric power measurement information to users.
In one embodiment, inputting sample user identity information and corresponding historical consulting information into a first power measurement information recommendation model comprises: converting the sample user identity information into a first word vector and converting the historical consultation information into a second word vector; the first word vector and the second word vector are input into a first power measurement information recommendation model.
In this embodiment, the terminal 102 may train the first power measurement information recommendation model by using the sample user identity information, the historical consultation information, and the corresponding sample power measurement information. The terminal 102 may input the processed sample user identity information and the historical consulting information into the first power measurement information recommendation model, for example, the terminal 102 may convert the sample user identity information into a first word vector and convert the historical consulting information into a second word vector, and input the first word vector and the second word vector into the first power measurement information recommendation model. The terminal 102 may perform word vector conversion through a preset pre-training model. For example, the terminal 102 may use a Word2vec pre-training model to convert words and phrases in the text message into vectors.
After the terminal 102 inputs the first word vector and the second word vector into the first power measurement information recommendation model, the model may be trained through a cross entropy loss function, which is defined as follows:
Figure BDA0003250155300000131
wherein D represents all sample data, y ∈ {0,1} represents whether the user consulted a certain question, for example, 0 represents that the user did not consult, and 1 represents that the user consulted; and p (x) is a probability value output by the network after passing through the sigmoid unit, and represents the degree of interest of a user in a certain problem or data. The terminal 102 may thus train the first power measurement information recommendation model based on the loss function.
Through the embodiment, the terminal 102 may train the first power measurement information recommendation model based on the sample user identity information and the historical consultation information after vector conversion, and may specifically train through a cross entropy loss function, so that a target power measurement information recommendation model for power measurement information recommendation may be obtained. The electric power measurement information recommendation method and the electric power measurement information recommendation device can accurately and reasonably recommend electric power measurement information to users.
In one embodiment, querying a target user database according to the user identifier and the user consultation information to obtain corresponding target user information includes: if the user database has user identity information corresponding to the user identification, obtaining target user information according to the user identity information corresponding to the user identification and historical consultation information corresponding to the user identity information; if the user database does not have the user identity information corresponding to the user identification, obtaining the historical consultation information which is most similar to the user consultation information in the user database, and obtaining the target user information according to the most similar historical consultation information and the user identity information corresponding to the most similar historical consultation information.
In this embodiment, the terminal 102 may obtain corresponding target user information from the target user database by using the user identifier and the user consultation information. The terminal 102 may query whether corresponding user identity information exists in the user database according to the user identifier, and if so, the terminal 102 may use the user identity information corresponding to the user identifier in the user database and the historical consultation information corresponding to the user identity information as the target user information. If the terminal 102 detects that the application identity information corresponding to the user identifier does not exist in the user database, the terminal 102 may query whether historical consultation information most similar to the user consultation information exists in the target user database according to the user consultation information, and if so, the terminal 102 may use the most similar historical consultation information and the user identity information corresponding to the most similar historical consultation information as the target user information.
The user identifier may be information such as a mobile phone number or a name of the user. The terminal 102 may perform query of the target user information based on the information such as the mobile phone number and name of the user. For example, the terminal 102 may search through whether the user exists in the target user database according to key fields such as name or mobile phone number, and if the information of the current user exists in the user information database, input the user information into a trained model, that is, the target power measurement information recommendation model; if the user information database does not contain the information of the current user, the terminal 102 may return the information of the consulted problem and the like existing in the database and the user information closest to the current user consulting problem, as the basic information of the current user, and transmit the information into the trained model. Therefore, the terminal 102 can transmit the retrieved information of the name or the mobile phone number of the user, the questions consulted once and the like into the trained model for prediction.
With the present embodiment, the problem that the current consultant user information is not in the user information database is considered, that is, the user is a new user. At this time, the system selects a user with the approximate information having the highest similarity with the current user information from the user information database, and takes the information as the basic information of the current user to perform subsequent recommendation prediction work. Therefore, the terminal 102 can obtain the data of the target power measurement information recommendation model input by the user based on the personal information of the user and the information consulted by the user, thereby realizing accurate and reasonable power measurement information recommendation for the user.
In one embodiment, as shown in fig. 5, fig. 5 is a flowchart illustrating a power measurement information recommendation method in another embodiment. The terminal 102 may recommend the power measurement information to the user by using a deep learning model based on the power measurement knowledge base and the user information database, i.e., the data in the target user database. The power measurement knowledge base is used for storing data related to power measurement. The user information database is used for storing basic information of users of the consultation questions, including names, mobile phone numbers, consulted questions, given answers and the like. The electric power measurement knowledge recommendation module based on the deep learning model comprises: based on the trained deep learning recommendation model. The method comprises the steps of inputting basic information of a user currently consulted and outputting five pieces of relevant information of power measurement with highest interest probability of the user.
Fig. 6 shows a specific flow, and fig. 6 is a schematic flow chart of a power measurement recommendation method in another embodiment. The method comprises the following steps: step S1, establishing a power measurement knowledge base, wherein the power measurement knowledge base comprises a plurality of power measurement information; step S2, establishing a user information database, namely the target user database, wherein the database stores the identity information of a plurality of users and the corresponding historical consultation information of the users; step S3, constructing a deep learning model to obtain the first power measurement information recommendation model to be trained; step S4, training a model by using data in a power measurement knowledge base and a user information database to obtain the target power measurement information recommendation model; step S5, searching the information of the user currently consulted in the user information database; step S6, transmitting the retrieved information of the current user into a trained deep learning model; and step S7, outputting the power measurement knowledge information interested by the current user by the model.
Through the embodiment, the terminal 102 can reasonably and accurately recommend the power measurement information to the user through the recommendation model based on the user identity information, the user consultation information and the target power measurement information.
It should be understood that although the various steps in the flowcharts of fig. 2-6 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 2-6 may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, which are not necessarily performed in sequence, but may be performed in turn or alternately with other steps or at least some of the other steps.
In one embodiment, as shown in fig. 7, there is provided an electricity measurement information recommendation apparatus, including: an obtaining module 500, a querying module 502, and a recommending module 504, wherein:
the obtaining module 500 is configured to obtain a user identifier and user consulting information of a user to be recommended in response to a power measurement information consulting instruction triggered by the user to be recommended.
A query module 502, configured to query a target user database according to the user identifier and the user consultation information, and obtain corresponding target user information; the target user information comprises user identity information and historical consultation information corresponding to the user to be recommended; the target user database stores the corresponding relation between a plurality of user identifications and historical consultation information.
The recommending module 504 is configured to input target user information into a target power measurement information recommending model, and obtain and display target power measurement information corresponding to a user to be recommended and output by the target power measurement information recommending model; the target power measurement information recommendation model is obtained based on a plurality of sample user identity information, a plurality of historical consultation information and a plurality of sample power measurement information training.
In one embodiment, the above apparatus further comprises: the construction module is used for acquiring a historical power measurement information consultation record corresponding to a user; acquiring user identity information of a user, historical consultation questions of the user and corresponding historical consultation answers from historical electric power measurement information consultation records; identifying power measurement information in the historical consultation questions and the historical consultation answers as historical consultation information; and obtaining a user database according to the user identity information and the historical consultation information.
In one embodiment, the above apparatus further comprises: the training module is used for acquiring a plurality of sample user identity information, a plurality of historical consultation information and a plurality of sample electric power measurement information; a group of sample user identity information and historical consulting information correspond to at least one sample power measurement information; acquiring a first power measurement information recommendation model to be trained, inputting sample user identity information and corresponding historical consultation information into the first power measurement information recommendation model, acquiring predicted power measurement information output by the first power measurement information recommendation model, and judging whether the similarity between the predicted power measurement information and the sample user identity information and the similarity between the predicted power measurement information and the sample power measurement information corresponding to the historical consultation information are larger than or equal to a preset similarity threshold value or not; if not, adjusting the first power measurement information recommendation model according to the similarity, and returning to the step of inputting the sample user identity information and the corresponding historical consultation information into the first power measurement information recommendation model; if so, ending the circulation, and taking the current first power measurement information recommendation model as a target power measurement information recommendation model.
In one embodiment, the training module is specifically configured to construct an embedding layer; the embedded layer is used for converting the sample user identity information into a first low-dimensional vector and converting the historical consultation information into a second low-dimensional vector; constructing an attention layer; the attention layer is used for identifying power measurement information in the first low-dimensional vector and the second low-dimensional vector through a multi-head attention mechanism; constructing a multilayer perceptron; the multi-layer sensing machine is used for identifying whether the power measurement information output by the attention layer is the power measurement information corresponding to the identity of the sample user, taking the power measurement information corresponding to the identity of the sample user as the predicted power measurement information and outputting the predicted power measurement information through a preset activation function; and obtaining a first power measurement information recommendation model according to the embedded layer, the attention layer and the multilayer perceptron.
In an embodiment, the training module is specifically configured to convert the sample user identity information into a first word vector and convert the historical consulting information into a second word vector; the first word vector and the second word vector are input into a first power measurement information recommendation model.
In an embodiment, the query module 502 is specifically configured to, if user identity information corresponding to a user identifier exists in the user database, obtain target user information according to the user identity information corresponding to the user identifier and historical consultation information corresponding to the user identity information; if the user database does not have the user identity information corresponding to the user identification, obtaining the historical consultation information which is most similar to the user consultation information in the user database, and obtaining the target user information according to the most similar historical consultation information and the user identity information corresponding to the most similar historical consultation information.
In an embodiment, the recommending module 504 is specifically configured to input the user identity information and the historical consulting information into the target power measurement information recommending model; acquiring a plurality of target power measurement information corresponding to a user to be recommended and output by a target power measurement information model and the prediction accuracy corresponding to each target power measurement information; and displaying a preset number of target power measurement information in the plurality of target power measurement information according to the prediction accuracy.
For specific limitations of the power measurement information recommendation device, reference may be made to the above limitations of the power measurement information recommendation method, which is not described herein again. All or part of the modules in the power measurement information recommendation device can be realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as shown in fig. 8. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless communication can be realized through WIFI, an operator network, NFC (near field communication) or other technologies. The computer program is executed by a processor to implement a power measurement information recommendation method. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.
Those skilled in the art will appreciate that the architecture shown in fig. 8 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, a computer device is provided, which includes a memory and a processor, wherein the memory stores a computer program, and the processor implements the power measurement information recommendation method when executing the computer program.
In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, the computer program, when executed by a processor, implementing the power measurement information recommendation method described above.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database or other medium used in the embodiments provided herein can include at least one of non-volatile and volatile memory. Non-volatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical storage, or the like. Volatile Memory can include Random Access Memory (RAM) or external cache Memory. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others.
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.

Claims (10)

1. A power measurement information recommendation method is characterized by comprising the following steps:
responding to a power measurement information consultation instruction triggered by a user to be recommended, and acquiring a user identifier and user consultation information of the user to be recommended;
inquiring a target user database according to the user identification and the user consultation information to acquire corresponding target user information; the target user information comprises user identity information and historical consultation information corresponding to the user to be recommended; the target user database stores corresponding relations between a plurality of user identifications and historical consultation information;
inputting the target user information into a target power measurement information recommendation model, and acquiring and displaying target power measurement information corresponding to the user to be recommended and output by the target power measurement information recommendation model; the target power measurement information recommendation model is obtained by training based on a plurality of sample user identity information, a plurality of historical consultation information and a plurality of sample power measurement information.
2. The method of claim 1, further comprising:
acquiring a historical power measurement information consultation record corresponding to a user;
acquiring user identity information of the user, historical consultation questions of the user and corresponding historical consultation answers from the historical electric power measurement information consultation record;
identifying the historical consultation question and the power measurement information in the historical consultation response as historical consultation information;
and obtaining the user database according to the user identity information and the historical consultation information.
3. The method of claim 1, further comprising:
acquiring a plurality of sample user identity information, a plurality of historical consultation information and a plurality of sample power measurement information; a set of the sample user identity information and historical consulting information corresponds to at least one of the sample power measurement information;
acquiring a first power measurement information recommendation model to be trained, inputting the sample user identity information and the corresponding historical consultation information into the first power measurement information recommendation model, acquiring predicted power measurement information output by the first power measurement information recommendation model, and judging whether the similarity between the predicted power measurement information and the sample user identity information as well as the similarity between the predicted power measurement information and the sample power measurement information corresponding to the historical consultation information is greater than or equal to a preset similarity threshold value;
if not, adjusting the first power measurement information recommendation model according to the similarity, and returning to the step of inputting the sample user identity information and the corresponding historical consultation information into the first power measurement information recommendation model;
if so, ending the circulation, and taking the current first power measurement information recommendation model as the target power measurement information recommendation model.
4. The method of claim 3, wherein obtaining the first power measurement information recommendation model to be trained comprises:
constructing an embedding layer; the embedding layer is used for converting the sample user identity information into a first low-dimensional vector and converting the historical consultation information into a second low-dimensional vector;
constructing an attention layer; the attention layer is used for identifying power measurement information in the first low-dimensional vector and the second low-dimensional vector through a multi-head attention mechanism;
constructing a multilayer perceptron; the multilayer perceptron is used for identifying whether the power measurement information output by the attention layer is power measurement information corresponding to the identity of a sample user, taking the power measurement information corresponding to the identity of the sample user as predicted power measurement information and outputting the predicted power measurement information through a preset activation function;
and obtaining the first power measurement information recommendation model according to the embedded layer, the attention layer and the multilayer perceptron.
5. The method of claim 3, wherein said inputting the sample user identity information and the corresponding historical consulting information into the first power metric information recommendation model comprises:
converting the sample user identity information into a first word vector and converting the historical consulting information into a second word vector;
inputting the first word vector and the second word vector into the first power measurement information recommendation model.
6. The method of claim 1, wherein the querying a target user database according to the user identifier and the user consulting information to obtain corresponding target user information comprises:
if the user database has user identity information corresponding to the user identification, obtaining the target user information according to the user identity information corresponding to the user identification and historical consultation information corresponding to the user identity information;
if the user database does not have the user identity information corresponding to the user identification, obtaining the most similar historical consultation information in the user database to the user consultation information, and obtaining the target user information according to the most similar historical consultation information and the user identity information corresponding to the most similar historical consultation information.
7. The method according to claim 1, wherein the inputting the target user information into a target power measurement information recommendation model, and obtaining and displaying target power measurement information corresponding to the user to be recommended output by the target power measurement information recommendation model, comprises:
inputting the user identity information and the historical consultation information into a target power measurement information recommendation model;
acquiring a plurality of target power measurement information corresponding to the user to be recommended and output by the target power measurement information model and the prediction accuracy corresponding to each target power measurement information;
and displaying a preset number of target power measurement information in the plurality of target power measurement information according to the prediction accuracy.
8. An electric power measurement information recommendation device, the device comprising:
the acquisition module is used for responding to a power measurement information consultation instruction triggered by a user to be recommended, and acquiring a user identifier and user consultation information of the user to be recommended;
the query module is used for querying a target user database according to the user identification and the user consultation information to acquire corresponding target user information; the target user information comprises user identity information and historical consultation information corresponding to the user to be recommended; the target user database stores corresponding relations between a plurality of user identifications and historical consultation information;
the recommendation module is used for inputting the target user information into a target power measurement information recommendation model, and acquiring and displaying target power measurement information corresponding to the user to be recommended and output by the target power measurement information recommendation model; the target power measurement information recommendation model is obtained by training based on a plurality of sample user identity information, a plurality of historical consultation information and a plurality of sample power measurement information.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 7.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.
CN202111043138.6A 2021-09-07 2021-09-07 Power measurement information recommendation method, device, computer equipment and storage medium Active CN113722601B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111043138.6A CN113722601B (en) 2021-09-07 2021-09-07 Power measurement information recommendation method, device, computer equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111043138.6A CN113722601B (en) 2021-09-07 2021-09-07 Power measurement information recommendation method, device, computer equipment and storage medium

Publications (2)

Publication Number Publication Date
CN113722601A true CN113722601A (en) 2021-11-30
CN113722601B CN113722601B (en) 2024-04-16

Family

ID=78682137

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111043138.6A Active CN113722601B (en) 2021-09-07 2021-09-07 Power measurement information recommendation method, device, computer equipment and storage medium

Country Status (1)

Country Link
CN (1) CN113722601B (en)

Citations (31)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2011081728A (en) * 2009-10-09 2011-04-21 Ntt Data Corp Device, method and program for extracting recommended content
US8751287B1 (en) * 2008-03-07 2014-06-10 Amazon Technologies, Inc. Generating item recommendations
CN104111938A (en) * 2013-04-18 2014-10-22 腾讯科技（深圳）有限公司 Information recommendation method and device
KR20150051505A (en) * 2013-11-04 2015-05-13 주식회사 티앤엠테크 Apparatus and method for recommending contract demand
US20160109495A1 (en) * 2014-10-17 2016-04-21 Florida Power And Light Company System and method for analyzing and monitoring smart meter network communications
CN106294497A (en) * 2015-06-09 2017-01-04 深圳市腾讯计算机系统有限公司 Information recommendation method and device
CN106485370A (en) * 2016-11-03 2017-03-08 上海智臻智能网络科技股份有限公司 A kind of method and apparatus of information prediction
US20180081883A1 (en) * 2012-06-01 2018-03-22 Tencent Technology (Shenzhen) Company Limited User recommendation method and system in sns community, and computer storage medium
EP3327650A1 (en) * 2016-11-29 2018-05-30 Samsung Electronics Co., Ltd. Method for recommending content and apparatus therefor
CN108763493A (en) * 2018-05-30 2018-11-06 深圳市思迪信息技术股份有限公司 A kind of recommendation method based on deep learning
US20180352091A1 (en) * 2017-06-01 2018-12-06 Adobe Systems Incorporated Recommendations based on feature usage in applications
US10244286B1 (en) * 2018-01-30 2019-03-26 Fmr Llc Recommending digital content objects in a network environment
WO2019070925A1 (en) * 2017-10-06 2019-04-11 Elsevier, Inc. Systems and methods for providing recommendations for academic and research entities
CN109885772A (en) * 2019-02-27 2019-06-14 上海优谦智能科技有限公司 The education content personalized recommendation system of knowledge based map
CN110287421A (en) * 2019-06-28 2019-09-27 北京金山安全软件有限公司 Information content recommendation method and device and electronic equipment
CN110309427A (en) * 2018-05-31 2019-10-08 腾讯科技（深圳）有限公司 A kind of object recommendation method, apparatus and storage medium
US20190319868A1 (en) * 2019-06-25 2019-10-17 Intel Corporation Link performance prediction technologies
CN110413877A (en) * 2019-07-02 2019-11-05 阿里巴巴集团控股有限公司 A kind of resource recommendation method, device and electronic equipment
US20200013237A1 (en) * 2017-03-31 2020-01-09 Honda Motor Co., Ltd. Maintenance notification system and method for controlling same, and non-transitory computer readable medium
CN110825957A (en) * 2019-09-17 2020-02-21 中国平安人寿保险股份有限公司 Deep learning-based information recommendation method, device, equipment and storage medium
CN111046286A (en) * 2019-12-12 2020-04-21 腾讯科技（深圳）有限公司 Object recommendation method and device and computer storage medium
CN111222044A (en) * 2019-12-31 2020-06-02 深圳Tcl数字技术有限公司 Information recommendation method and device based on emotion perception and storage medium
CN111402013A (en) * 2020-06-04 2020-07-10 成都晓多科技有限公司 Commodity collocation recommendation method, system, device and storage medium
CN111815050A (en) * 2020-07-07 2020-10-23 北京嘀嘀无限科技发展有限公司 Model training method, model training device, state prediction method, state prediction device, electronic equipment and storage medium
CN112035611A (en) * 2020-08-28 2020-12-04 康键信息技术（深圳）有限公司 Target user recommendation method and device, computer equipment and storage medium
CN112100221A (en) * 2019-06-17 2020-12-18 腾讯科技（北京）有限公司 Information recommendation method and device, recommendation server and storage medium
CN112381112A (en) * 2020-10-16 2021-02-19 华南理工大学 User identity recognition method and system based on multi-mode item set of user data
CN112765453A (en) * 2020-12-31 2021-05-07 平安资产管理有限责任公司 Content recommendation method and device, computer equipment and storage medium
CN112836115A (en) * 2019-11-25 2021-05-25 浙江大搜车软件技术有限公司 Information recommendation method and device, computer equipment and storage medium
CN112991009A (en) * 2021-03-08 2021-06-18 北京嘀嘀无限科技发展有限公司 Object recommendation method and device, electronic equipment and storage medium
CN113239173A (en) * 2021-06-09 2021-08-10 深圳集智数字科技有限公司 Method and device for processing question and answer data, storage medium and electronic equipment

Patent Citations (31)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8751287B1 (en) * 2008-03-07 2014-06-10 Amazon Technologies, Inc. Generating item recommendations
JP2011081728A (en) * 2009-10-09 2011-04-21 Ntt Data Corp Device, method and program for extracting recommended content
US20180081883A1 (en) * 2012-06-01 2018-03-22 Tencent Technology (Shenzhen) Company Limited User recommendation method and system in sns community, and computer storage medium
CN104111938A (en) * 2013-04-18 2014-10-22 腾讯科技（深圳）有限公司 Information recommendation method and device
KR20150051505A (en) * 2013-11-04 2015-05-13 주식회사 티앤엠테크 Apparatus and method for recommending contract demand
US20160109495A1 (en) * 2014-10-17 2016-04-21 Florida Power And Light Company System and method for analyzing and monitoring smart meter network communications
CN106294497A (en) * 2015-06-09 2017-01-04 深圳市腾讯计算机系统有限公司 Information recommendation method and device
CN106485370A (en) * 2016-11-03 2017-03-08 上海智臻智能网络科技股份有限公司 A kind of method and apparatus of information prediction
EP3327650A1 (en) * 2016-11-29 2018-05-30 Samsung Electronics Co., Ltd. Method for recommending content and apparatus therefor
US20200013237A1 (en) * 2017-03-31 2020-01-09 Honda Motor Co., Ltd. Maintenance notification system and method for controlling same, and non-transitory computer readable medium
US20180352091A1 (en) * 2017-06-01 2018-12-06 Adobe Systems Incorporated Recommendations based on feature usage in applications
WO2019070925A1 (en) * 2017-10-06 2019-04-11 Elsevier, Inc. Systems and methods for providing recommendations for academic and research entities
US10244286B1 (en) * 2018-01-30 2019-03-26 Fmr Llc Recommending digital content objects in a network environment
CN108763493A (en) * 2018-05-30 2018-11-06 深圳市思迪信息技术股份有限公司 A kind of recommendation method based on deep learning
CN110309427A (en) * 2018-05-31 2019-10-08 腾讯科技（深圳）有限公司 A kind of object recommendation method, apparatus and storage medium
CN109885772A (en) * 2019-02-27 2019-06-14 上海优谦智能科技有限公司 The education content personalized recommendation system of knowledge based map
CN112100221A (en) * 2019-06-17 2020-12-18 腾讯科技（北京）有限公司 Information recommendation method and device, recommendation server and storage medium
US20190319868A1 (en) * 2019-06-25 2019-10-17 Intel Corporation Link performance prediction technologies
CN110287421A (en) * 2019-06-28 2019-09-27 北京金山安全软件有限公司 Information content recommendation method and device and electronic equipment
CN110413877A (en) * 2019-07-02 2019-11-05 阿里巴巴集团控股有限公司 A kind of resource recommendation method, device and electronic equipment
CN110825957A (en) * 2019-09-17 2020-02-21 中国平安人寿保险股份有限公司 Deep learning-based information recommendation method, device, equipment and storage medium
CN112836115A (en) * 2019-11-25 2021-05-25 浙江大搜车软件技术有限公司 Information recommendation method and device, computer equipment and storage medium
CN111046286A (en) * 2019-12-12 2020-04-21 腾讯科技（深圳）有限公司 Object recommendation method and device and computer storage medium
CN111222044A (en) * 2019-12-31 2020-06-02 深圳Tcl数字技术有限公司 Information recommendation method and device based on emotion perception and storage medium
CN111402013A (en) * 2020-06-04 2020-07-10 成都晓多科技有限公司 Commodity collocation recommendation method, system, device and storage medium
CN111815050A (en) * 2020-07-07 2020-10-23 北京嘀嘀无限科技发展有限公司 Model training method, model training device, state prediction method, state prediction device, electronic equipment and storage medium
CN112035611A (en) * 2020-08-28 2020-12-04 康键信息技术（深圳）有限公司 Target user recommendation method and device, computer equipment and storage medium
CN112381112A (en) * 2020-10-16 2021-02-19 华南理工大学 User identity recognition method and system based on multi-mode item set of user data
CN112765453A (en) * 2020-12-31 2021-05-07 平安资产管理有限责任公司 Content recommendation method and device, computer equipment and storage medium
CN112991009A (en) * 2021-03-08 2021-06-18 北京嘀嘀无限科技发展有限公司 Object recommendation method and device, electronic equipment and storage medium
CN113239173A (en) * 2021-06-09 2021-08-10 深圳集智数字科技有限公司 Method and device for processing question and answer data, storage medium and electronic equipment

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
SHUN LI: "A personalized electricity tariff recommender system based on advanced metering infrastructure and collaborative filtering", 《 INTERNATIONAL JOURNAL OF ELECTRICAL POWER & ENERGY SYSTEMS》, vol. 113, pages 403 - 410, XP085724302, DOI: 10.1016/j.ijepes.2019.05.042 *
ZHENZHEN XU: "Cross-domain item recommendation based on user similarity", 《COMPUTER SCIENCE AND INFORMATION SYSTEMS》, vol. 13, no. 2, pages 359 - 373 *
张敏: "推荐算法及其在电力营销中的应用", 《中国优秀硕士学位论文全文数据库 工程科技》, pages 138 - 1460 *

Also Published As

Publication number Publication date
CN113722601B (en) 2024-04-16

Similar Documents

Publication Publication Date Title
US20120317102A1 (en) Ranking expert responses and finding experts based on rank
CN111680219A (en) Content recommendation method, device, equipment and readable storage medium
CN112364204B (en) Video searching method, device, computer equipment and storage medium
CN109710739B (en) Information processing method and device and storage medium
EP3475883A1 (en) Dynamically managing artificial neural networks
CN110401545B (en) Chat group creation method, chat group creation device, computer equipment and storage medium
CN111192153B (en) Crowd relation network construction method, device, computer equipment and storage medium
CN113505204A (en) Recall model training method, search recall device and computer equipment
Boahen et al. Detection of compromised online social network account with an enhanced knn
CN108885637A (en) Personage is mild-natured related
CN112182174A (en) Business question-answer knowledge query method and device, computer equipment and storage medium
Dai et al. Sequential behavior prediction based on hybrid similarity and cross-user activity transfer
Deng et al. Neural variational collaborative filtering with side information for top-K recommendation
CN111311455B (en) Examination information matching method, examination information matching device, computer equipment and storage medium
CN113869063A (en) Data recommendation method and device, electronic equipment and storage medium
Taheri et al. DeepMovRS: a unified framework for deep learning-based movie recommender systems
CN113722601B (en) Power measurement information recommendation method, device, computer equipment and storage medium
CN115114448B (en) Intelligent multi-mode fusion power consumption inspection method, device, system, equipment and medium
Gao et al. An efficient protein homology detection approach based on seq2seq model and ranking
CN111615178B (en) Method and device for identifying wireless network type and model training and electronic equipment
Eichinger et al. Affinity: A system for latent user similarity comparison on texting data
CN113888216A (en) Advertisement information pushing method and device, electronic equipment and storage medium
Wang et al. A matrix factorization based framework for fusion of physical and social sensors
CN117473070B (en) Multi-channel application method of intelligent robot, intelligent robot and storage medium
Narayanan et al. Towards open ended and free form visual question answering: Modeling vqa as a factoid question answering problem

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
CB02 Change of applicant information

Country or region after: China

Address after: Room 86, room 406, No.1, Yichuang street, Zhongxin Guangzhou Knowledge City, Huangpu District, Guangzhou City, Guangdong Province

Applicant after: Southern Power Grid Digital Grid Research Institute Co.,Ltd.

Applicant after: CHINA SOUTHERN POWER GRID Co.,Ltd.

Address before: Room 86, room 406, No.1, Yichuang street, Zhongxin Guangzhou Knowledge City, Huangpu District, Guangzhou City, Guangdong Province

Applicant before: Southern Power Grid Digital Grid Research Institute Co.,Ltd.

Country or region before: China

Applicant before: CHINA SOUTHERN POWER GRID Co.,Ltd.

CB02 Change of applicant information
TA01 Transfer of patent application right

Effective date of registration: 20240313

Address after: Room 86, room 406, No.1, Yichuang street, Zhongxin Guangzhou Knowledge City, Huangpu District, Guangzhou City, Guangdong Province

Applicant after: Southern Power Grid Digital Grid Research Institute Co.,Ltd.

Country or region after: China

Address before: Room 86, room 406, No.1, Yichuang street, Zhongxin Guangzhou Knowledge City, Huangpu District, Guangzhou City, Guangdong Province

Applicant before: Southern Power Grid Digital Grid Research Institute Co.,Ltd.

Country or region before: China

Applicant before: CHINA SOUTHERN POWER GRID Co.,Ltd.

TA01 Transfer of patent application right
GR01 Patent grant